A flexibly mounted aircraft engine may under certain conditions experience a self-excited whirling instability involving a coupling between the gyroscopic and aerodynamic forces acting on the propeller, and the inertial, elastic, and damping forces contributed by the power plant, nacelle, and wing. This phenomenon has been called autoprecession, or whirl instability. An experimental investigation was made in the Langley transonic dynamics tunnel at Mach numbers below 0.3 to study some of the pertinent parameters influencing the phenomenon. These parameters included propeller rotational speed, stiffness of the power-plant assembly in the pitch and yaw planes and the ratio of pitch stiffness to yaw stiffness, structural damping of the power-plant assembly in the pitch and yaw planes, simulated fuel load in the wings, and the location and number of autoprecessing powerplant assemblies. A large dynamic-aeroelastic model of a four-engine turboprop transport airplane mounted on a vertical rod in a manner which provided several limited body degrees of freedom was used in the investigation. It was found that the boundary for autoprecession decreased markedly with Increasing proreduction of power-plant stiffness and/or damping, and to a lesser degree decreased with reduction of simulated fuel load in the wings. peller rotational speed generally lowered the autoprecession boundary. This effect was more pronounced as the stiffness was increased. An inboard power plant was found to be more susceptible to autoprecession than an outboard one. Combinations in which two or more power plants had the same level of reduced stiffness resulted in autoprecession boundaries considerably lower than that of a single power plant with the same level of reduced stiffness.